a dissection of the teashirt and tiptop genes reveals a novel ......oof 1 a dissection of the...

1

2

3Q1

4

5

6789101112131415161718192021

44

45

46

47

48

49

50

51

52

53

54

55

56

57

58

59

60

61Q2

Developmental Biology xxx (2011) xxx–xxx

YDBIO-05477; No. of pages: 12; 4C: 4, 5, 6, 7, 9, 10

Contents lists available at SciVerse ScienceDirect

Developmental Biology

j ourna l homepage: www.e lsev ie r .com/deve lopmenta lb io logy

F

A dissection of the teashirt and tiptop genes reveals a novel mechanism for regulatingtranscription factor activity

Rhea R. Datta, Brandon P. Weasner, Justin P. Kumar⁎Department of Biology, Indiana University, Bloomington, IN 47405, USA
O
⁎ Corresponding author.E-mail address: [email protected] (J.P. Kumar).

0012-1606/$ – see front matter © 2011 Published by Eldoi:10.1016/j.ydbio.2011.09.030

Please cite this article as: Datta, R.R., et al., Afactor activity, Dev. Biol. (2011), doi:10.101

Oa b s t r a c t
a r t i c l e i n f o
22

23

24

25

26

27

28

29

30

31

32

33

Article history:Received for publication 13 August 2010Revised 2 September 2011Accepted 27 September 2011Available online xxxx

Keywords:TeashirtTiptopHomothoraxCtBPRetinal determinationCell proliferation

34

35

36

37

38

39

40

41

ECTED P

RIn the Drosophila eye the retinal determination (RD) network controls both tissue specification and cellproliferation. Mutations in network members result in severe reductions in the size of the eye primordiumand the transformation of the eye field into head cuticle. The zinc-finger transcription factor Teashirt (Tsh)plays a role in promoting cell proliferation in the anterior most portions of the eye field as well as in inducingectopic eye formation in forced expression assays. Tiptop (Tio) is a recently discovered paralog of Tsh. It isdistributed in an identical pattern to Tsh within the retina and can also promote ectopic eye development.In a previous study we demonstrated that Tio can induce ectopic eye formation in a broader range of cellpopulations than Tsh and is also a more potent inducer of cell proliferation. Here we have focused onunderstanding the molecular and biochemical basis that underlies these differences. The two paralogsare structurally similar but differ in one significant aspect: Tsh contains three zinc finger motifs whileTio has four such domains. We used a series of deletion and chimeric proteins to identify the zinc fingerdomains that are selectively used for either promoting cell proliferation or inducing eye formation. Ourresults indicate that for both proteins the second zinc finger is essential to the proper functioning of theprotein while the remaining zinc finger domains appear to contribute but are not absolutely required.Interestingly, these domains antagonize each other to balance the overall activity of the protein. This appearsto be a novel internal mechanism for regulating the activity of a transcription factor. We also demonstrate thatboth Tsh and Tio bind to C-terminal Binding Protein (CtBP) and that this interaction is important for promotingboth cell proliferation and eye development. And finally we report that the physical interaction that has beendescribed for Tsh and Homothorax (Hth) do not occur through the zinc finger domains.

© 2011 Published by Elsevier Inc.

4243
R
62

63

64

65

66

67

68

69

70

71

72

73

74

75Q376Q4Q577

78

Q4, Q5

UNCO

RIntroduction

The developing compound eye of Drosophila is an excellent systemto study how the rates of cell proliferation and tissue specificationare balanced. These two processes are controlled, in part, by the activityof a group of genes that is collectively referred to as the retinaldetermination (RD) network.

The genes that comprise this network include the Pax genes eyeless(ey), twin of eyeless (toy), eyegone (eyg) and twin of eyegone (toe)(Aldaz et al., 2003; Czerny et al., 1999; Jang et al., 2003; Quiring etal., 1994), the Six family homologs sine oculis (so) and optix (Cheyetteet al., 1994; Seimiya and Gehring, 2000; Serikaku and O'Tousa, 1994),the protein tyrosine phosphatase eyes absent (eya) (Bonini et al.,1993), a homolog of the Ski/Sno proto-oncogene dachshund (dac)(Mardon et al., 1994), the Nl kinase nemo (nmo) (Braid and Verheyen,2008; Choi and Benzer, 1994), the Meis homolog homothorax (hth)(Gonzales-Crespo et al., 1998; Pai et al., 1997; Pichaud and Casares,

79

80

81

82

sevier Inc.

dissection of the teashirt and6/j.ydbio.2011.09.030

2000; Rauskolb et al., 1995), the pipsqueak transcription factors distalantenna (dan) and distal antenna related (danr) (Curtiss et al., 2007) aswell as the zinc finger transcription factors teashirt (tsh) and tiptop(tio) (Bessa et al., 2009; Datta et al., 2009; Laugier et al., 2005; Panand Rubin, 1998). Mutationswithinmany of these genes lead to drastic,if not complete, reductions in the developing eye primordium. Thenumber of cells in the eye field is dramatically lowered and anysurviving cells adopt a head cuticle fate (Bonini et al., 1993; Cheyetteet al., 1994; Jang et al., 2003; Mardon et al., 1994; Quiring et al.,1994). Nearly all of the genes that belong to the RD network arealso capable of redirecting the fate of non-retinal tissues towardsan eye in forced expression assays (Bessa et al., 2009; Bonini et al.,1997; Braid and Verheyen, 2008; Curtiss et al., 2007; Czerny et al.,1999; Datta et al., 2009; Halder et al., 1995; Pan and Rubin, 1998;Seimiya and Gehring, 2000; Shen and Mardon, 1997; Weasner etal., 2007). Thus this network is a key regulatory of both tissuespecification and cell proliferation.

tsh was first identified in a screen of enhancer traps that reportedexpression of genes crucial for embryogenesis (Fasano et al., 1991).Upon activation by several Hox and pair-rule genes (Core et al.,1997; Mathies et al., 1994; McCormick et al., 1995), Tsh, in turn,

tiptop genes reveals a novel mechanism for regulating transcription

http://dx.doi.org/10.1016/j.ydbio.2011.09.030

mailto:[email protected]


http://www.sciencedirect.com/science/journal/00121606


T

83

84

85

86

87

88

89Q6790

91

92

93

94

95

96

97

98

99

100

101

102

103

104

105

106

107

108

109

110

111

112

113

114

115

116

117

118Q8119

120

121

122

123

124

125

126

127

128

129

130

131

132

133

134

135

136

137

138

139

140

141

142

143

144

145

146

147

148

149

150

151

152

153

154

155

156

157

158

159

160

161

162

163

164

165

166

167

168

169

170

171

172

173

174

175

176

177

178

179

180

181

182

183

184

185

186

187

188

189

190

191

192

193

194

195

196

197

198

199

200

201

202

203

204

205

206

207

Q6, Q7

2 R.R. Datta et al. / Developmental Biology xxx (2011) xxx–xxx

UNCO

RREC

then cooperates with Hox proteins to repress head specifying genesand to establish trunk identities (Andrew et al., 1994; Coiffier et al.,2008; Roder et al., 1992). Subsequently, tsh has been shown to playroles in several additional tissues including the embryonic salivaryglands (Henderson et al., 1999), larval legs (Erkner et al., 1999;2002; Wu and Cohen, 2000) and wings at both larval and adult stages(Soanes et al., 2001; Sun et al., 1995; Wu and Cohen, 2000). The firstinsight into a role for tsh in eye development came from a screen of P[lacZ] lines that showed a non-uniform pattern of w+ expression inthe adult retina (Sun et al., 1995). tsh is expressed and regulatesboth patterning and proliferation ahead of the morphogenetic furrow.A trimeric complex containing Tsh, Ey andHomothorax (Hth) represseseye development, via transcriptional repression of several retinaldetermination genes including so, eya and dac in the most anteriorportions of the eye disc (Bessa et al., 2002). This allows for thecontinued proliferation of retinal progenitor cells, which is mediatedby another trimeric complex containing Tsh, Hth and Yorki (Yki), adownstream target of the Hippo tumor suppressor pathway (Lopesand Casares, 2009; Peng et al., 2009). It is not yet known if repressionof RD genes or the promotion of cell proliferation requires Tsh todirectly interact with DNA. These activities occur in context specificsituations as patterning defects due to the loss of tsh are onlyobserved in the dorsal half of the retinawhile tshdependent proliferationdefects predominant within the ventral half (Singh et al., 2002). Onepotential explanation for the region specific phenotypes is that theloss of tsh may be masked by the activity of additional factors.

A prime candidate is the tio gene, which is a paralog of tsh and isexpressed in an identical pattern to tsh in the retina (Bessa et al.,2009; Datta et al., 2009; Laugier et al., 2005). In support of bothgenes being redundant to each other, complete loss-of-function tiomutants are homozygous viable with no discernable effects on thestructure of the eye, a phenotype very similar to certain tsh mutantclones (Laugier et al., 2005; Pan and Rubin, 1998; Singh et al., 2002).This apparent redundancy is based on amutual repressionmechanismin which loss of either gene leads to increased and compensatoryexpression of the other factor (Bessa et al., 2009; Laugier et al.,2005). Indeed, more pronounced defects are observed when tshlevels are knocked down in a tio null mutant background (Peng et al.,2009). It is not clear, however, if this repression is direct or thoughintermediate factors. Despite this redundancy, forced expression assaysusing several hundred GAL4 lines revealed differences between the twoparalogs. These assays indicated that tio is a more proficient inducer ofectopic eyes and can stimulate higher levels of cell proliferation thantsh (Datta et al., 2009). Clues to what underlies the differences inactivitiesmay be grounded in the presence of an extra zincfingerwithinTio (Laugier et al., 2005) and/or in potential amino acid differenceswithin homologous zinc finger domains (Datta et al., 2011). Tsh hasbeen shown to bind to DNA and at least one direct target, modulo(mod), has been identified (Alexandre et al., 1996). To date, directtargets of Tio have not been identified and it is not clear if the bindingof Tsh to DNA is due to the zinc finger motifs or if it is dependentupon co-factors.

Here we report the results of a molecular dissection of the tsh andtio paralogous genes. We have made a series of deletion constructs inwhich individual zinc fingers have been removed. Additionally, wegenerated a chimeric protein in which the unique zinc finger domainwithin Tio was added to Tsh. These deletion and chimeric proteinswere assayed for their ability to induce ectopic eye formation andcell proliferation in forced expression assays. We demonstrate herethat these two activities are absolutely dependent upon the presenceof the second zinc finger. Interestingly, the first zinc finger contributesto both eye formation and cell proliferation while the third zinc fingerfunctions to suppress both activities. This balancing of activitieswithin a single transcription factor may represent a novel mechanismfor regulating transcription. Our results also indicate that, contrary toour a priori expectations, the unique fourth zinc finger in Tio is not

Please cite this article as: Datta, R.R., et al., A dissection of the teashirt andfactor activity, Dev. Biol. (2011), doi:10.1016/j.ydbio.2011.09.030

ED P

RO

OF

solely responsible for the differences between the Tsh and Tio.We go on to show that the zinc fingers do not mediate the physicalinteractions that have been reported for Tsh and Hth. However,Tsh and Tio dependent patterning and proliferation in the retinarequire interactions with C-terminal Binding Protein (CtBP), atranscriptional co-repressor. Finally, we demonstrate that a Tsh/Tioortholog from the flour beetle, Tribolium castaneum, is capable forinducing ectopic eyes in Drosophila. This result suggests that themolecular and biochemical mechanisms by which Tsh and Tio regulateeye formation are evolutionarily conserved.

Materials and methods

Fly strains

We used the following 19 fly stocks: (1) dpp-GAL4; (2) yweyFLP;FRT82B Ubi-GFP; (3) hsFLP22; act5C>y+>GAL4, UAS-GFP; (4) FRT82BCtBPS7De-10; (5) UAS-CtBP; (6) yw M[vas-int.Dm]ZH-2A; PBac[y+-attP-3B]VK00033; (7) UAS-Tsh FL; (8) UAS-TshΔZn1; (9) UAS-TshΔZn2;(10) UAS-TshΔZn3; (11) UAS-TshΔPLDLS; (12) UAS-Tio FL; (13) UAS-TioΔZn1; (14) UAS-TioΔZn2; (15) UAS-TioΔZn3; (16) UAS-TioΔZn4;(17) UAS-TioΔPLDLS; (18) UAS-Tsh/Tio Zn4; (19) UAS-Tc Tsh/Tio (Tri-bolium). All UAS-tsh and UAS-tio variants were generated using thephiC31 integration system (Bischof and Basler, 2008; Bischof et al.,2007; Venken et al., 2006).

Deletion and chimeric molecules

The full-length Tsh protein is 954 amino acids in length and can bedivided into seven segments: the N-terminal (NT), zinc finger 1 (Zn1),first linker segment (L1), zinc finger 2 (Zn2), second linker segment(L2), zinc finger 3 (Zn3) and the C-terminal tail (CT). The NT segmentcontains residues 1–355, Zn1 contains residues 356–378; L1 containsresidues 379–467, Zn2 contains residues 468–490, L2 contains residues491–534, Zn3 contains residues 535–557 and the CT segment containsresidues 558–954. The PLDLS domain is located within the NT segmentand consists of residues 188–192. The Zn1 deletion (TshΔZn1) containsamino acids 1–355 fused to residues 379–954, the Zn2 deletion(TshΔZn2) contains amino acids 1–467 fused to residues 491–954, theZn3 deletion (TshΔZn3) contains amino acids 1–534 fused to residues558–954, the CtBP binding site deletion (TshΔPLDLS) contains aminoacids 1–187 fused to residues 193–954.

The full-length Tio protein is 1025 amino acids in length and can bedivided into nine segments: the N-terminal (NT), zinc finger 1 (Zn1),first linker segment (L1), zinc finger 2 (Zn2), second linker segment(L2), zinc finger 3 (Zn3), third linker segment (L3), zinc finger 4(Zn4) and the C-terminal tail (CT). The NT segment contains residues1–318, Zn1 contains residues 319–341; L1 contains residues 342–427,Zn2 contains residues 428–450, L2 contains residues 451–500, Zn3contains residues 501–523, L3 contains residues 524–927, Zn4 containsresidues 928–949 and the CT segment contains residues 950–1025. ThePLDLS domain is locatedwithin theNT segment and consists of residues187–191. The Zn1deletion (TioΔZn1) contains amino acids 1–318 fusedto residues 342–1025, the Zn2 deletion (TioΔZn2) contains amino acids1–427 fused to 451–1025, the Zn3 deletion (TioΔZn3) contains aminoacids 1–500 fused to residues 524–1025, the Zn4 deletion (TioΔZn4)contains amino acids 1–927 fused to residues 950–1025). The CtBPbinding site deletion (TioΔPLDLS) contains amino acids 1–186 fusedto residues 192–1025. The Tsh/Tio Zn4 chimera contains amino acids1–954 of Tsh fused to residues 928–1025 of Tio.

Clonal analysis

CtBP loss-of-function clones were generated using the directedmosaic approach with yweyFLP; FRT82B CtBPS7De-10/FRT82B Ubi-GFPflies. Over-expressing clones of either tsh or tio variants were generated



T

208

209

210

211

212

213

214

215

216

217

218

219

220

221

222

223

224

225

226

227

228

229

230

231

232

233

234

235

236

237

238

239

240

241

242

243

244

245

246

247

248

249

250

251

252

253

254

255

256

257

258

259

260

261

262

263

264

265

266

267

268

269

270

271

272

273

274

275

276

277

278

279

280

281

282

283

284

285

286

287

288

289

290

291

292

293

294

295

296

297

298

299

300

301

302

303

304

305

306

307

308

309

310

311

Q9313

314

315

316

317

318

319

320

321

Q10323

324

325

3R.R. Datta et al. / Developmental Biology xxx (2011) xxx–xxx

UNCO

RREC

by heat shocking hsFLP22; act5C>y+>GAL4, UAS-GFP, UAS-tsh variantor hsFLP22; act5C>y+>GAL4, UAS-GFP, UAS-tio variant for 30 min at34 °C between 60 and 84 h after embryos were deposited.

Immunostaining and antibodies

Imaginal discs were dissected from developing larvae and werefixed in 4% paraformaldehyde in PBS for 45 min at room temp. Tissueswere blocked in a solution containing 10% goat serum and 0.1% Triton,incubated with primary antibodies at room temperature overnight,washed in PBS+0.1% Triton, incubated with Cy5, TRITC or FITCconjugated secondary antibodies (Jackson Immunoresearch) for 2–4 hat room temp, washed in PBS+0.1% Triton andmounted in Vectashield(Vector). Images were taken with a Zeiss Axioplan II fluorescentmicroscope with Apotome using Axiovision 4.6 software. Adult flieswith ectopic eyes were frozen at −80 °C for 20 min and images weretaken with a Zeiss Discovery light microscope and MRc color camera.

Primary antibodies used were rat anti-Elav (1:100, DSHB), mouseanti-Dac (1:5, DSHB), mouse anti-Eya (1:5, DSHB), rabbit anti-Tsh(1:3000, Stephen Cohen), guinea pig anti-CtBP (1:500, YutakaNibu); rabbit anti-PH3 (1:1000, Abcam). All secondary antibodiesused in this study were obtained from Jackson Laboratories and aredonkey anti-mouse TRITC (1:20), donkey anti-rat FITC (1:20), donkeyanti-rat Cy5 (1:20), goat anti-rat FITC (1:20), goat anti-rat TRITC (1:20),goat anti-rabbit TRITC (1:20) goat anti-rabbit FITC (1:20). F-actin wasvisualized by TRITC conjugated phalloidin.

Quantifications: ectopic eye size and frequency, cell proliferation

All UAS-GAL4 crosses were carried out at 25 °C. dpp-GAL4 femaleswere crossed to either UAS-tsh variant or UAS-tio variant males andwere allowed to lay eggs for 45 min. Individual eggs were placed inmicrocentrifuge tubes containing fly media and allowed to age for96 h. This eliminates variations in cell proliferation and apoptosis thatare due to differences in population densities. Larvae of the appropriategenotype were then dissected and eye-antennal discs prepared asdescribed above. Ectopic eyes were identified using rat anti-Elavwhile dividing cells were marked with rabbit anti-PH3. The size of theectopic eye and the zones of increased proliferation were calculatedusing NIH ImageJ software. Each disc was quantified three times(technical replicates) with 30 biological replicates for each genotype.Ectopic eye frequencies were also calculated for pupae and adult flies. pvalueswere calculated for cell proliferation/tissue growthquantifications.

Yeast 2-hybrid screen

The ProQuest Yeast 2-Hybrid System (Invitrogen) was used toidentify proteins that physically associate with Tsh and Tio. A libraryof cDNAs from Drosophila third instar larvae was cloned into thepDEST-22 vector, which contains the GAL4 activation domain (preyplasmid). Full-length tsh and tio cDNAs were cloned into the pDEST-32 vector, which contains theGAL4DNAbinding domain (bait plasmid).Yeast MaV203 cells were transformed with the prey library and eachbait plasmid. Interactions between prey proteins and either Tsh or Tiowere identified by the activation of three reporters (UAS-lacZ, UAS-HIS3 and UAS-URA3). Plasmids from positive clones were isolated andsequenced.

Immunoprecipitation

Kc167 cells were transfected with combinations of the following 15plasmids (0.4 μg each): (1) mt-GAL4; (2) UAS-tsh FL-MYC; (3) UAS-tshΔZn1-MYC; (4) UAS-tsh ΔZn2-MYC; (5) UAS-tsh ΔZn3-MYC; (6) UAS-tsh ΔPLDLS-MYC; (7) UAS-tio FL-MYC; (8) UAS-tio ΔZn1-MYC; (9)UAS-tio ΔZn2-MYC; 10) UAS-tio ΔZN3-MYC; (11) UAS-tio ΔZn4-MYC;(12) UAS-tio ΔPLDLS-MYC; (14) UAS-CtBP-HA; (15) UAS-Hth-HA by


ED P

RO

OF

using Qiagen Effectene Transfection Reagent and induced after 24 hwith 1 mMCuSO4. For immunoprecipitation studies nucleiwere isolatedand nuclear proteins isolated using NE-PER Nuclear and CytoplasmicExtraction Reagents (Thermo)with protease inhibitors. After conductinga pre-clear step (to eliminate non-specific interactions) with protein Gagarose beads the supernatant was incubated with 5 μl of eitheranti-HA or anti-MYC antibodies at 4 °C for overnight followed by anadditional 3 h incubation at 4 °C with protein G agarose beads. Thesamples were spun and washed three times in RIPA buffer+25%acetonitrile, boiled in SDS loading buffer+β-mercaptoethanol andresolved on 10% SDS-PAGE gels. All proteins were visualized onimmunoblots using mouse anti-HA (Santa Cruz) and mouse anti-MYC(Santa Cruz) primary antibodies, goat anti-mouse HRP secondaryantibodies (Jackson Immunoresearch) and the SuperSignal WestPico Chemiluminescent Substrate (Thermo Scientific). Blots wereprocessed and imaged using a Chemi-Doc XRS+from BioRad.

Confirmation of genomic integration of UAS lines

Genomic integration all UAS-tsh variant and UAS-tio variantconstructs were confirmed using PCR of genomic DNA from a singlefly of each transformant line. Two fragments of 134 and approximately450 bp are observed in the yw M[vas-int.Dm]ZH-2A; PBac[y+-attP-3B]VK00033 stock. Successful integration yields PCR fragments of 227 and454 bp.

Confirmation of protein production

All tsh and tio variant cDNAs were cloned into a modified Gatewayexpression vector containing an in-frame HA epitope tag. S2 cellswere transfected with 0.4 μg plasmids containing mt-GAL4 and eitherUAS-tsh variant-V5 or UAS-tio variant-V5 by using Qiagen EffecteneTransfection Reagent and induced after 24 h with 1 mM CuSO4. 3 mlof cells were pelleted and then resuspended and lysed in lysis buffer.The lysate was boiled in SDS loading buffer+β-mercaptoethanol andresolved on 10% SDS-PAGE gels. All proteins were visualized onimmunoblots using mouse anti-V5 (Santa Cruz) and mouse anti-Actinprimary antibodies, goat anti-mouseHRP secondary antibodies (JacksonImmunoresearch) and the SuperSignal West Pico ChemiluminescentSubstrate (Thermo Scientific).

Results

We, as well as others, have previously shown that in forcedexpression assays both Tsh and Tio are capable of inducing ectopic eyeformation and in promoting tissue proliferation (Bessa et al., 2009;Datta et al., 2009; Pan and Rubin, 1998; Singh et al., 2002). Interestingly,despite the functional redundancy that exists between the twoparalogs, Tio appears to be a more potent inducer of eye formation andtissue growth (Bessa et al., 2009; Datta et al., 2009). These differencescould be due to the presence of a unique zinc finger motif in Tio or dueto amino acid sequence differences between the homologous zinc fingerdomains (Datta et al., 2011; Laugier et al., 2005). In this paper we set outto determine the relative contributions that each zincfingermakes to theoverall function of both proteins and in the process determine if thedifferences between Tsh and Tio are attributable to either distinctionsamongst the homologous zinc fingers or to the unique zinc finger that isfound only in Tio. We also set out to establish if any of the zinc fingerdomains mediate the formation of the Tsh–Hth complex (Peng et al.,2009). And finally we ascertained if physical interactions withC-terminal Binding Protein, a known transcriptional co-repressor (Nibuet al., 1998a, 1998b), which have been documented for Tsh (Saller etal., 2002) and predicted for Tio (Laugier et al., 2005) are necessarypromoting eye development and tissue growth.

In order to answer these questions we generated several reagentsfor a molecular dissection study. The first set of reagents includes a



326

327

328

329

330

331

332

333

334

335

336

337

338

339

340

341

342

343

344

345

346

347

348

349

350

351

352

353

354

355

356

357

358

359

360

361

362

363

364

365

366

367

368

369

370

371


series of Tsh and Tio variants in which individual zinc finger domainsor the CtBP binding site (PLDLS) have been deleted (Fig. 1A). We alsogenerated a chimeric protein in which the unique zinc finger domainwithin Tio was fused to the full-length Tsh protein (Fig. 1A). And finallywe have generated a construct in which the single Tsh/Tio orthologfrom the flour beetle, Tribaolium castaneum can be expressed in flies(Fig. 1A). Each of these constructs was placed under UAS control andusing the phiC31 recombination system integrated into the samelocation within the genome to control for expression levels and thusallowing for comparisons across genotypes. All insertions events wereconfirmed using PCR (Fig. 1B; see Materials and methods). An epitopetagged version of each construct was transfected into S2 cells in orderto document that each protein can be generated and is stable(Fig. 1C). Note that of all the constructs the only one that we couldnot detect is the Tribolium Tsh/Tio protein (Fig. 1C). It is possible thatthe Tribolium Tsh/Tio protein is itself unstable in S2 cells or that theaddition of the epitope tag renders it unstable. We do however observethat its expression in the antenna induces ectopic eyes and higher levelsof cell proliferation (see below).

Antagonistic activity of Tsh and Tio Zn fingers balance ectopic eye formation

Forced expression of either Tsh or Tio in the dpp expression domaininduces ectopic eyes in just the antennal disc (Fig. 2A–C; Datta et al.,

UNCO

RRECT

Fig. 1. Schematic of Tsh and Tio constructs, genomic integration and protein confirmation. (Astudy. The location of the PLDLS and Zn domains are found at the top of the panel. (B) PCRthat for each construct the expected bands of 227 and 454 bp are obtained. (C) A western bstable proteins for all constructs except for Tribolium Tsh/Tio.


RO

OF

2009) at frequencies of 92% and 98% respectively (Fig. 2D,E). We thenassayed the effects of removing each zinc finger on these frequenciesin larval discs. For both proteins the first zinc finger contributes to butis not absolutely required for the formation of ectopic eyes as expres-sion of TshΔZn1 and TioΔZn1 induces ectopic eyes at 43% and 86%(Fig. 2D,E). In contrast, we find that the second zinc finger is absolutelyrequired for ectopic eye formation as expression of TshΔZn2 andTioΔZn2 supports ectopic eyes in only 1% and 4% of flies (Fig. 2D,E).The putative requirement for this particular zinc finger in both proteinsis further supported by the fact that of the three homologous zinc fingerpairs Zn2 is the most closely related in Tsh and Tio (Datta et al., 2011).The fact that we can detect stable TshΔZn2 and TioΔZn2 proteins inS2 cells (Fig. 1C) suggests that the proteins are also likely to be stablein the antennal disc. Surprisingly, expression of variants in which thethird zinc finger has been deleted (TshΔZn3, TioΔZn3) lead to a statisti-cally significant increase in ectopic eye frequencies that approaches100% for both proteins (Fig. 2D,E). We interpret this to mean that Zn3plays an inhibitory role and thus counteracts that activity of both Zn1and Zn2 (Fig. 8A).

Deletion of the unique fourth zinc finger (TioΔZn4) decreases thefrequency of ectopic eye formation to 22% (Fig. 2E) suggesting that it,like Zn1, contributes to but is not required for the formation of ectopiceyes (Fig. 8A). Previously, we had shown the while tio can induceectopic eyes when driven with a set of eight GAL4 drivers, tsh can

ED P

) A schematic diagram of all Tsh and Tio constructs that were generated and used in thisanalysis to confirm the proper location of each construct after phiC31 integration. Notelot showing expression of each construct in S2 cells. Note that we were able to detect



TD P

RO

OF

372

373

374

375

376

377

378

379

380

381

382

383

384

385

386

387

388

389

390

391

392

393

394

395

396

397

398

399

400

401

402

403

404

405

406

407

408

409

410

411

412

413

414

415

416

417

418

419

420

421

422

423

424

425

426

427

428

429

430

431

432

433

434

435

436

437

Fig. 2. Frequencies of ectopic eyes induced by Tsh and Tio variants. (A) A confocal image of a third instar disc containing an ectopic eye within the antenna. Anterior is to the right.(B,C) Light microscope images of adult flies containing ectopic eyes (arrows). (D,E) Graphs documenting the frequencies of ectopic eye formation induced by each Tsh and Tiovariant.


UNCO

RREC

only induce eye development when expressed in a smaller subset ofexpression domains (Datta et al., 2009). In terms of topologicalrange of GAL4 driver expression patterns, expression of TioΔZn4 isstill capable of inducing ectopic eyes with all eight GAL4 driversthat were previously reported for Tio and not with just the smallersubset that was reported with Tsh (data not shown). We next deter-mined if Zn4 on its own was capable of increasing the frequencyand topological range of ectopic eyes generated by Tsh. To do thiswe expressed within the antennal disc a chimeric protein in whichZn4 of Tio was fused to the full-length Tsh protein (Tsh/Tio Zn4).This chimeric protein induced ectopic eyes in 100% of larvae(Fig. 2D) further suggesting that Zn4 plays an instructive role inpromoting eye development. Expression of Tsh/Tio Zn4was, however,only capable of supporting eye development when expressed withinthe smaller and more limited set of GAL4 drivers that were previouslyreported for full-length Tsh (data not shown). Based on the totality ofthese results we can order the zinc finger domains in terms of theirimportance for promoting eye formation. For Tsh it appears that therelationship Zn2>Zn1 applies while for Tio the relationship isZn2>Zn4>Zn1. In both cases Zn3 functions as a counterbalance tothe other zinc finger domains (Fig. 8A).

The flour beetle, Tribolium castaneum, contains a single Tsh/Tio likegene (Shippy et al., 2008). Since its protein structure is more similar tothat of Drosophila Tio we, a priori, expected that expression ofTribolium Tsh/Tio in the fly antenna would induce ectopic eyesat a rate comparable to either Tsh or Tio. However, we obtainedfrequencies of only 25%, which is significantly lower than that ofeither Drosophila ortholog. We conclude that although the frequency ofectopic eye formation for Tribolium Tsh/Tio is lower than that of theDrosophila orthologs, the fact that ectopic eyes can be induced at all issupportive of a model in which both Tsh/Tio mediated protein–proteinand protein–DNA interactions have been conserved between these twospecies despite nearly 300 million years of evolutionary separation.Functional conservation amongst Tsh/Tio family members likely


E

extends to mammalian systems as expression of at least one mouseTsh gene (mTsh3)within the dpp domain induces ectopic eye formationin the antenna.

Activities of Tsh and Tio Zinc fingers regulate ectopic eye size

Tsh is known to play a role in maintaining a pool of dividingprogenitor cells within the normal retina (Lopes and Casares,2009; Peng et al., 2009; Singh et al., 2002). In a previous reportwe had noted that the ectopic eyes induced by expression ofboth Tsh and Tio were accompanied by an over-proliferation ofthe antennal disc (Datta et al., 2009). The size of the ectopic eyesthat were induced by tio expression is substantially larger thanthose generated by tsh. During the course of this study we observedthat the removal of individual zinc fingers not only affects thefrequency at which ectopic eyes were generated but the deletionsalso affect the amount of tissue overgrowth and the final size of theadult ectopic retinas. Using our ectopic eye assay and our zinc fingerdeletion set we attempted to determine the contribution that eachzinc finger makes to Tsh dependent cell proliferation. We alsoattempted to determine if the differences in Tsh and Tio inducedovergrowth are attributable to one or more of the zinc fingerdomains. In order to accomplish this we stained dpp-GAL4/UAS-tshand dpp-GAL4/UAS-tio full-length and variant(s) eye-antennaldiscs with an antibody that detects the pan-neuronal protein Elavand then used NIH ImageJ software to measure the area within theantennal disc that is occupied by the newly induced retina(Fig. 3A–E).

Our measurements here support our initial observations that theectopic eyes induced by Tio are larger than those that are inducedby Tsh (Fig. 3A,B,E; p=3.68E−07). Removal of Zn1 and Zn3 fromTsh leads, surprisingly, to an increase in the size of the ectopic eyes(Fig. 3C,E; TshΔZn1 p=0.002, TshΔZn3 p=1.50619E−07). In contrast,we do not observe any statistically significant differences in the size of



ORRECTED P

RO

OF

438

439

440

441

442

443

444

445

446

447

448

449

450

451

452

453

454

455

456

457

458

459

460

461

462

463

464

465

466

467

468

469

470

471

Fig. 3. Contribution of Each Zn finger to ectopic eye size. (A–D) Confocal images of third instar eye discs. Genotypes of each disc are at the bottom right of each panel. Visualizedmolecules are listed at the top right of each panel. Anterior is to the right. (E) Graph documenting the size ectopic eyes that are induced by each Tsh and Tio variant. Measurementswere taken with ImageJ software.


UNCthe ectopic retinas that are induced by the removal of the homologous

zinc fingers from Tio (Fig. 3E; TioΔZn1 p=0.872, TioΔZn3 p=0.109).The presence of Zn2 is not only required to induce retinal development(Fig. 2D,E), but its removal from either Tsh or Tio expectedly also leadsto a severe reduction in the size of the eyes that are produced (Fig. 3D,Ered bars; TshΔZn2 p=5.62E−06, TioΔZn2 p=7.42E−10). This resultsuggests that both Tsh and Tio require Zn2 to promote cell proliferationand tissue specification. Interestingly, while Zn4 of Tio contributes sig-nificantly to the frequency of ectopic eyes (Fig. 2E) its removal has noeffect on the size of the ectopic retina (Fig. 3E; TioΔZn4 p=0.21;Fig. 3E). However, its addition to Tsh leads to a rather dramatic increasein the size of the retina when compared to the effects of Tsh alone(Fig. 3E; Tsh/Tio Zn4 p=6.97518E−06). From these results we con-clude that all three zinc fingers of Tsh contribute to the determinationof final organ size with activity of individual zinc fingers counter balan-cing each other (Fig. 8B). In contrast, all functional activity related to Tiodependent cell proliferation is embedded just within Zn2 (Fig. 8B). It isunclear why, on the one hand, removal of Zn4 from Tio has no effect on


tissue sizewhile, on the other hand, its addition to Tsh has an amplifyingeffect.

We extended this assay to Tribolium Tsh/Tio in order to determineif the role in cell proliferation has ancient origins. As its structure ismore similar withDrosophila Tshwe expected that its ability to inducetissue growthwouldmimic that ofDrosophila Tio. However, the size ofthe ectopic eyes that are induced by Tribolium Tsh/Tio are significantlysmaller than those induced by Tio (Fig. 3E; p=0.019) and are notsignificantly different than those induced by Tsh (Fig. 3E; p=0.084).This result suggests that while Tribolium Tsh/Tio does not induceectopic eyes to the frequency of either Tsh or Tio, its role in regulatingthe size of the retina is more similar to Tsh than Tio.

Restrictions on the induction of tissue growth by Tsh and Tio

The ability of either Tsh or Tio to induce ectopic eyes is restrictedto certain subpopulations of the antennal disc (Bessa et al., 2009;Datta et al., 2009; Pan and Rubin, 1998). We were interested in



T

472

473

474

475

476

477

478

479

480

481

482

483

484

485

486

487

488

489

490

491

492

493

494

495

496

497

498

499

500

501

502

503

504

505

506

507

508

509

510

511

512

513

514

515

516

517

518

519

520

521

522

523

524

525

526

527

528

529

530

531

532

533

534

535

536

537

538

539

540

541


determining whether the ability of each protein to induce tissue pro-liferation is similarly restricted within the antennal disc. We generatedover-expression Tsh FL or Tio FL clones randomly throughout theeye-antennal disc and assayed for the presence or absence of tissueovergrowths. To make a quantitative measurement of tissue growthwe counted the number of PH3 positive cells (which are in M phase ofthe cell cycle) within the clone and then normalized these counts toeliminate differences in clone size. In contrast to neutral GFP clones,which do not induce cell proliferation (data not shown), we observedtissue overgrowths for both Tsh FL and Tio FL. While Tio is capableof inducing proliferation throughout the entire disc (Fig. 4B, Table 1),tissue overgrowth caused by Tsh expression was limited to the ventralcompartment (Fig. 4A, Table 1). Compartment specific effects areobserved in the eye disc when Tsh is either lost through mutationor over-expressed (Singh et al., 2002; 2004).

One potential explanation for the compartment specific effects inboth the eye and the antenna is that Tsh may physically interactwith factors that have restricted distribution patterns. As zinc fingerscan mediate protein–protein interactions we set out to determine ifany of the zinc fingers of Tsh and Tio contribute to the location oftissue outgrowth production and would thus serve as candidates formediating tissue specific interactions between Tsh/Tio and potentialcofactors. Removal of Zn1 from either Tsh or Tio did not alter the loca-tion of ectopic eye formation (Fig. 4C,D; Table 1) although the size ofthe outgrowths induced by Tio ΔZ1 is significantly smaller than thatTio FL (Table 1). Consistent with the absolute requirement of Zn2 forinducing ectopic eyes, expression of either TshΔZn2 or TioΔZn2 failedto induce cell proliferation within either the eye or antennal disc(Fig. 4E,F; Table 1). Zn3 of Tsh appears to be important for limitingthe ability of Tsh to promote cell proliferation as expression ofTshΔZn3 is now capable of inducing tissue overgrowth in the dorsalcompartment a phenotype that we do not observe with Tsh FL(Fig. 4A,G arrow; Table 1). On the other hand, removal of Zn3 fromTio does not have any discernable effect on the location of inducedcell proliferation (Fig. 4H, Table 1). The effects of Zn4 on tissue over-growth are very similar to that seen with ectopic eye size. Just as

UNCO

RREC

Fig. 4. Location of tissue outgrowths induced by Tsh and Tio variants. (A–K) Confocal imageVisualized molecules are listed at the bottom left of panel A. Anterior is to the right. Dottedectopic tissue growths. Blue indicates location of ectopic tissue growth that is induced by bothat is induced specifically by the Tio full-length molecule. Green indicates location that bo


ED P

RO

OF

expression of TioΔZn4 does not significantly influence the size ofthe ectopic retina, there is no effect on the location of induced cellproliferation (Fig. 4I, Table 1). We observe that expression of Tsh/Tio Zn4 is also not sufficient to induce tissue overgrowths in thedorsal compartment (Fig. 4J, Table 1). Expression of Tribolium Tsh/Tioinduces cell proliferation to a similar extent that we see for Tio both interms of size of the tissue outgrowth and in location throughout the an-tennal disc (Fig. 4K, Table 1). This contrasts to the effects we see onthe initiation of eye development in which Tribolium Tsh/Tio inducesectopic eyes at a frequency that is significantly lower than that ofeither Tsh or Tio (Fig. 2D,E). These results might suggest thatroles of Tsh and Tio in promoting tissue growth are quite ancientand were present prior to the Tribolium/Drosophila split.

Tsh and Tio interactions with Hth are not mediated by the zinc fingerdomains

Tsh is contained within at least two biochemical complexes thatare required to promote cell proliferation. The first complex, whichcontains Tsh, Hth and Ey, functions to repress eye development byshutting off transcription of the so, eya and dac genes in the mostanterior portions of the eye disc (Bessa et al., 2002). This allowsfor a second complex, which contains Tsh, Hth and Yki, to promoteproliferation of retinal cell precursor cells (Peng et al., 2009). Sinceseveral of our deletion constructs, particularly those that removeZn2, are unable to promote cell proliferation we set out to determineif the interactions between Tsh and Hth are mediated by any ofthe individual zinc fingers. We also were interested in determiningif Tio can form a complex with Hth and if any of its zinc fingerdomains are functional relevant to this interaction. A plasmid containinga full-length Hth clone was co-transfected into Kc167 cells along withplasmids containing either full-length or zinc finger deletion versionsof Tsh and Tio. We isolated nuclear protein fractions and immunopreci-pitated complexes containing Hth. These complexes were then probedfor the presence of the full-length and zinc finger deletions of eitherTsh or Tio. As expected Tsh FL and Tio FL biochemically interact with

s of third instar eye discs. Genotypes of each disc are at the bottom right of each panel.lines indicate examples of tissue outgrowths. (L) A summary diagram of the location ofth Tsh and Tio full-length molecules. Yellow indicates location of ectopic tissue growthth Tsh and Tio induce cell proliferation and ectopic eye formation.



T

542

543

544

545

546

547

548

549

550

551

552

553

554

555

556

557

558

559

560

561

562

563

564

565

566

567

568

569

Q11571

572

573

574

Q12576

577

578

579

580

581

582

583

584

585

586

587

588

589

590

591

592

593

Table 1t1:1

Location of tissue proliferation, quantification and confidence values.t1:2t1:3 Construct Dorsal location Dorsal quantification PH3/unit area Dorsal confidence Ventral location Ventral quantification Ventral confidence

t1:4 Tsh FL (−) 0.000956629 p=0.0003* (+) 0.003479516 p=0.0156*t1:5 Tsh ΔZn1 (−) 0.001383797 p=0.1285** (+) 0.003256138 p=0.6556**t1:6 Tsh ΔZn2 (−) 0.000550406 p=0.2451** (−) 0.000284591 p=3.8299E−07**t1:7 Tsh ΔZn3 (+) 0.003062254 p=6.4860E−05** (+) 0.003516603 p=0.9398**t1:8 Tsh/Tio Zn4 (−) 0.0017417 p=0.0004** (+) 0.003235934 p=0.5694**t1:9 Tio FL (+) 0.002564655 p=0.0003** (+) 0.00510712 p=0.0156**t1:10 Tio ΔZn1 (+) 0.001585035 p=0.0307* (+) 0.002050515 p=0.0014*t1:11 Tio ΔZn2 (−) 0.000262221 p=4.7409E−05* (−) 0.000269899 p=4.4531E−07*t1:12 Tio ΔZn3 (+) 0.003741986 p=0.0333* (+) 0.003287827 p=0.0249*t1:13 Tio ΔZn4 (+) 0.002406333 p=0.8061* (+) 0.003183437 p=0.0022*t1:14 Tc Tsh/Tio (+) 0.002596489 p=0.0001** (+) 0.002865935 p=0.2432**t1:15 p=0.9367* p=0.0082*

*compared to Tio FL, **compared to Tsh FL.t1:16

Q17


REC

Hth (Fig. 5, lanes 1 and 5). The removal of individual zinc fingers did notdisrupt the binding to Hth as each protein variant was detected afterimmunoprecipitation with Hth (Fig. 5, lanes 2–4 and 6–9). Our mockimmunoprecipitations (using a generic IgG) failed to bring down anyof the tagged proteins. We therefore conclude that Hth interactswith both Tsh and Tio through one of the non-zinc finger protein seg-ments. At this point we cannot rule out the possibility that the zincfinger domains do not interact with other factors such as Ey and Ykior whether they primarily interact with DNA target sequences. Wecan say with some certainty, however, that Zn2, which appears tobe the most critical zinc finger domain within Tsh and Tio, contributesto the activity of both proteins independently of Hth.

Tsh and Tio function through binding to CtBP

Tsh genes in both flies and vertebrates are thought to function astranscriptional repressors. During embryonic fly development theloss of tsh results in the ectopic expression of the Hox gene labial(lab) while over-expression eliminates modulo (mod) transcription(Alexandre et al., 1996; Roder et al., 1992). In the developing eye,Tsh is part of a complex that represses the transcription of severalretinal determination genes (Bessa et al., 2002). Additionally, lossof either tsh or tio is compensated by the ectopic expression of theother paralog (Bessa et al., 2009). What are the mechanisms bywhich Tsh and Tio repress transcription? At least one mechanisminvolves physical interactions with C-terminal Binding Protein (CtBP),a known transcriptional co-repressor (Nibu et al., 1998a,b). Drosophila

UNCO

R

Fig. 5. Interactions between Tsh, Tio and Hth are not mediated by the zinc finger domainthrough the non-zinc finger domains. The nuclear lysate rows demonstrate that each proteproteins are not non-specifically immunoprecipitated. The last row indicates that Hth was imthat both full-length and zinc finger deletion Tsh and Tio proteins interact specifically with


ED P

RO

OF

Tsh and Tio as well as the three murine Tsh homologs all contain aCtBP interaction motif, which appears to be important for Tshdependent repression of target genes during embryogenesis(Fig. 6A; Laugier et al., 2005; Manfroid et al., 2004; Saller et al.,2002). For instance, a complex containing Brinker (Brk), Tsh andCtBP function together to repress Ultrabithorax expression in themidgut (Saller et al., 2002) while the removal of the CtBP interactiondomain prevents Tsh from repressing embryonic mod expression(Manfroid et al., 2004).

Since Tio contains a PLDLS motif we set out to determine if Tio canalso form a biochemical complex with CtBP. We performed a yeast2-hybrid screen in which Tio was used as bait to screen a third instarlibrary that was enriched with cDNAs from the eye-antennal imaginaldiscs. Of the 50 clones that were identified in the screen 36 of themcorresponded to CtBP (data not shown). We were able to immuno-precipitate Tsh–CtBP and Tio–CtBP complexes from Kc167 cells(Fig. 6B lanes 1 and 3) thereby confirming that a portion of theirtranscriptional repressive activities is dependent upon CtBP. Todetermine if the induction of ectopic eyes and/or the promotionof cell proliferation is due to Tsh–CtBP and Tio–CtBP repressivecomplexes we expressed both TshΔPLDLS and TioΔPLDLS in thedeveloping antenna via dpp-GAL4. In both cases, the percentage ofectopic eyes was reduced dramatically. Expression of TshΔPLDLSinduced ectopic eyes in 11.5% (3/26) of screened animals whileectopic retinas were seen in only 12.8% (5/39) of animals in whichTioΔPLDLS was expressed. The ectopic eyes were very small withmost only containing a handful of ommatidia (data not shown). It

s. Western blots demonstrating that interactions between Hth and Tsh and Tio occurin is made and stable in Kc167 cells. The negative control lanes demonstrate that themunoprecipitated in each experiment. The second to last row (immunoblot) indicatesHth.



T

OO

F

594

595

596

597

598

599

600

601

602

603

604

605

606

607

608

609

610

611

612

613

614

615

616

617

618

Q13Q14

621

622

623

624

625

626

627

628

629

630

631

632

633

634

635

636

637

638

639

640

641

642

643

644

645

646

647

648

649

650

651

652

653

654

655

656

657

658

659

660

661

662

663

664

665

666

667

668

669

670

671

672

673

674

Fig. 6. CtBP interacts with Tsh and Tio and is expressed in the developing eye. (A) N-terminal sequences of Tsh and Tio. The PLDLS domain is highlighted in red. (B,C) Western blotand co-IP from S2 cells demonstrating that Tsh binds to CtBP through the PLDLS domain. (D,E) Western blot and co-EP from S2 cells demonstrating that Tio binds to CtBP throughthe PLDLS domain. (F–H) Confocal images of wild type third instar eye-antennal disc stained with antibodies directed against Elav (green) and CtBP (purple). The arrow indicatesarea of CtBP expression that overlaps with that of Tsh and Tio. Anterior is to the right.


UNCO

RREC

should be noted that both constructs are produced and are stable incell culture (Fig. 6B lanes 2 and 4 whole cell lysate). From theseresults we conclude that the PLDLS domain and thus the interactionwith CtBP is required for tissue growth. However, while the PLDLSdomain is important for Tsh and Tio dependent induction of ectopiceyes, it is not absolutely required since both proteins can induceectopic eyes even when the PLDLS domain has been removed.One possible explanation is that Tsh and Tio may interact withother transcriptional co-repressors such as Groucho (Gro). Such amechanism has been shown to exist for Brk, which can bind toboth CtBP and Gro (Hasson et al., 2001). Similarly, Knirps (Kni) canrepress transcriptional targets in the embryo by non-CtBP dependentinteractions (Keller et al., 2000). Alternatively, CtBP may still be ableto interact with both Tsh and Tio either through a different interactionmotif or through a bridging protein. We have some evidence that thismay be the case: both TshΔPLDLS and TioΔPLDLS protein variantscontinue to bind CtBP in cell culture (Fig. 6B lanes 2 and 4). As neitherprotein is found with the mock immunoprecipitations (using genericIgG antibodies) we are confident that these results are biologicallyrelevant and are not experimental artifacts. The existence of a bridgingmolecule has been suggested to bring CtBP in contact with Giant (Gt:Nibu and Levine, 2001). And the interactions between CtBP and itsbinding partners are not solely dependent upon the possession of aPXDLS consensus sequences surrounding secondary and tertiarystructures make interactions with CtBP considerably complex (Molloyet al., 2001). It should be noted that our results conflict with those ofManfroid et al., 2004, which show that the deletion of the PLDLSmotif abrogates binding to Tsh. We attribute the differences betweenour results to experimental techniques. Manfroid et al. used an invitro GST fusion system while our immunoprecipitations are done inKc167 cells. It is possible that a bridging protein is normally presentin Kc167 cells. If that is the case then Tsh and Tio would still interactwith CtBP despite the removal of the PLDLS domain. These types ofproteins would not be expected to be present within the GST fusionsystem.

In any event we wanted to determine the expression pattern inthe developing retina and determine what, if any, effect does removalof CtBP have on eye formation. CtBP appears to be expressed in allcells within the developing eye-antennal disc. It is enriched aheadof the morphogenetic furrow (Fig. 6C–E, arrow) where both Tsh andTio are normally expressed. We generated CtBP loss-of-function clonesin the retina and demonstrate that dac expression is de-repressed in


ED P

R

cells that normally express tsh and tio (Fig. 7A–D). Additionally, themorphogenetic furrow appears to accelerate through CtBP mutanttissue (Fig. 7E–H). This phenotype is very similar to that seen wheneither extramacrochaetae (emc) or ultraspiracle (usp) are removed(Brown et al., 1995; Zelhof et al., 1997). It is not clear if the advance-ment of furrow is directly due to the increased levels of Dac proteinor if CtBP directly regulates the Hh and Dpp signaling cascades.From these results we can however conclude that Tsh and Tio functionas repressors through their interaction with CtBP and that thesecomplexes play important roles in promoting both normal and ectopiceye development as well as in inducing tissue proliferation.

Discussion

In Drosophila, tsh and tio are paralogous genes that arose from anancient duplication event (Shippy et al., 2008). A growing body ofevidence has suggested that these genes play redundant roles inthe developing eye. For example, both genes are expressed in identicalpatterns within the developing retina (Bessa et al., 2009; Datta et al.,2009; Pan and Rubin, 1998). Internal loss-of-function tsh clonesoften display no overt phenotype (Pan and Rubin, 1998; Singh et al.,2002) while null mutants of tio are completely viable with no visibleeye defects (Laugier et al., 2005). Indeed, effects on growth controlare only revealed when both genes are simultaneously removed(Peng et al., 2009). Furthermore, while Tsh alone has been shown todirectly bind DNA (Alexandre et al., 1996) both proteins are thoughtto mutually repress each other's expression and thus are thought toboth be part of transcriptional repressive complexes (Bessa et al.,2002).

However, while these genes appear to play similar roles in thedeveloping eye there is evidence to suggest that they do influenceboth eye development and tissue proliferation to varying degrees.For example, while null Tio mutants have no visible eye phenotype,dorsal margin clones of tsh are transformed into head cuticle whileventral margin clones show overgrowth phenotypes (Laugier et al.,2005; Singh et al., 2002; 2004). Additionally, in forced expressionassays Tio appears to be a more proficient inducer of both ectopiceyes and cell proliferation (Datta et al., 2009). In this report we haveset out to understand the biochemical mechanisms that underlie func-tional differences between the zinc finger transcription factors Tshand Tio, to ascertain if any of the zinc fingers mediate interactions



TD P

RO

OF

675

676

677

678

679

680

681

Q15683

684

685

686

687

688

689

690

691

692

693

694

695

696

697

698

699

700

701

702

703

704

705

706

707

708

709

710

711

712

713

714

715

716

Fig. 7. Loss of CtBP leads to de-repression of dac and an acceleration of the furrow. (A–H) Confocal images of third instar larval eye discs. Visualized molecules are listed at thebottom right of each panel. (A,D) The arrows indicate areas of dac de-repression within CtBP clones. The arrowheads indicate area of normal dac expression in wild type tissue.(G,H) The arrows denote the acceleration of the furrow within CtBP clones. The arrowheads denote the normal furrow in wild type tissue. Anterior is to the right.

Fig. 8. A novel internal mechanisms for regulating transcription factor function. (A) Aschematic that summarizes the relative contributions that each Zn makes towards theprocesses of retinal development. Note that for Tsh Zn3 counteracts the effects of theother zinc fingers. Also note that such an internal balancing act does not exist for Tio.These differences may explain why Tio is a more potent inducer of ectopic eyes thanTsh. (B) A schematic that summarizes the relative contributions that each Znmakes towards the processes of cell proliferation. Note that while Zn1 and Zn3counterbalance the activity of Zn2 within Tsh no such competing mechanism existsfor Tio.


UNCO

RREC

with Hth and to determine if interactions with CtBP are required foreye specification and cell proliferation.

To investigate the first questionwe focused on themajor structuraldifference that exists between these two zinc finger transcriptionfactors: Tsh contains three such domains while Tio possesses fourmotifs and is thus more similar to the mouse and basal insect Tsh/Tioproteins (Caubit et al., 2000; Fasano et al., 1991; Laugier et al., 2005;Manfroid et al., 2004; Shippy et al., 2008). We generated a series ofdeletion constructs in which individual zinc fingers were removedfrom either Tsh or Tio. These constructs were then used in forcedexpression assays to assess the relative requirements for eachDNA binding domain in promoting eye development and inducingtissue proliferation. In a companion experiment we also generateda chimeric protein in which the fourth (and unique) zinc fingerfrom Tio was fused to the full-length Tsh protein. For comparisonthe effects that the deletion and chimeric constructs had on ectopiceye development and tissue growth were measured against themore ancient Tribolium Tsh/Tio protein. We find that in the case ofTsh, Zn1 and Zn2 are required for both the promotion of eye develop-ment and tissue growth. Countering the activities of these twodomains is Zn3, which appears to play a role in suppressing both pro-liferation and differentiation. Our dissection of Tio revealed a similarmechanism for regulating its activity during eye induction andretinal differentiation: Zn1, Zn2 and Z4 act to promote retinaldifferentiation while Zn3 appears to inhibit this process. To ourknowledge this balancing act represents a novel mechanism forregulating the activity of a transcription factor (Fig. 8A,B).

We were unable to uncover a mechanism that would account forthe higher proficiency in inducing ectopic eyes. Removal of Zn4from Tio (Tio ΔZn4) or its addition to Tsh (Tsh/Tio Zn4) did notalter the specificity with which Tio or Tsh can induce ectopic eyesnor did manipulations of any other zinc finger for that matter (datanot shown). That topological difference may reside within theremaining non-conserved protein segments and may involve interac-tions with different sets of binding proteins. The data supporting the


Esecond half of this hypothesis is somewhat mixed. Our previousefforts to identify binding factors using a genetic screen failed toyield paralog specific binding partners (Datta et al., 2009). Onthe other hand our yeast two-hybrid screen did identify a numberof potential paralog specific interacting proteins (data not shown).Further biochemical studies and in vivo functional assayswill be requiredto confirm or reject this line of reasoning.



T

717

718

719

720

721

722

723

724

725

726

727

728

729

730

731

732

733

734

735

736

737

738

739

740

741

742

743

744

745

746

747

748

749

750

751

752

753

754

755

756

757

758

759

760

761

762

763

764

765

766

767

768

Q16

770

771

772

773

774

775

776

777

778779780781782783784785786787788789790791792793794795796797798799800801802803804805806807808809810811812813814815816817818819820821822823824825826827828829830831832833834835836837838839840841842843


NCO

RREC

Our analysis of Tio revealed an interesting difference that mayprovide a mechanistic explanation for the higher proficiency of Tioin inducing tissue proliferation. While Zn1, Zn2 and Zn4 are allrequired to promote tissue proliferation, Zn3 appears to effectneither the location of tissue overgrowths nor the size of the inducedectopic eye. Without the use of an inhibitory zinc finger Tio has threesuch motifs being used to promote tissue growth while Tsh has onlytwo such motifs being used for this same purpose (Fig. 8B). Wepropose that the use of higher numbers of zinc fingers may allowfor Tio to form more stable protein–nucleic acid complexes thathave longer resident times on genes that directly affect the cell cycle.

It is also possible that each of the zinc fingers has different bindingpartners and/or transcriptional binding sites. It was shown that Tshand Hth act together, along with Yki and Ey to promote anterior eyedisc proliferation and suppress eye development (Bessa et al., 2002;Peng et al., 2009). We had hypothesized that Hth might bind to Tshand Tio through Zn2 to fulfill these functions. However, results fromour immunoprecipitation experiments suggest that the interactionsbetween Hth and both Tsh and Tio occur through the non-zinc fingerdomains. Additionally, Tsh and Tio may have different targets in thedorsal and ventral regions of the antenna and this may allow forincreased proliferation. And finally, each zinc finger might bind todistinct binding sites. Genomic and biochemical analyses will berequired to fully investigate these possible mechanistic options.

The tight regulation of RD genes is presumably linked to geneticand protein interactions, which is then related to gene structure.Different functional domains of Tsh and Tio are able to affect differentregions of the eye antennal disc when over-expressed. While mostdevelopmental genes are highly pleiotropic, it is especially intriguingthat Tsh and Tio are able to carry out all of their functions with thesame C2H2 zinc finger domain. We have previously shown thatthere is differential selection acting along the lengths of the genesand that relaxed selection acting on specific motifs contributes tochanges in function in highly pleiotropic gene (Datta et al., 2011). Itis possible that a coding sub-functionalization event has taken place,where one domain within Tsh and Tio acts as a repressor and theother acts as a promoter of eye development. A comparison with thePax6 family can be made. Eyg acts as a repressor of eye development,while Ey promotes eye development. The duplication event giving riseto Tsh/Tiomay have occurred at the same time as the Pax6/Pa65a split,allowing for the same amount of time for nucleotide substitution andsequence evolution. The only difference here is that the division offunction has taken place between genes in the Pax family, while sub-functionalization may have taken place within a gene itself.

And finally, the binding of Tsh and Tio to the transcriptionalco-repressor CtBP appears crucial for both proteins as removal of thisdomain interferes with the ability of Tsh and Tio to induce ectopiceyes and promote proliferation (data not shown). Furthermore, theloss of CtBP expression ahead of the furrow relieves the repression ofdac expression by Tsh (Fig. 7A–D). These results indicate that CtBP is amember of transcriptional repressive complexes that include Tsh andTio and plays a critical role within the retinal determination network.

844845846847848849850

U

Uncited reference

Soanes and Bell, 2001
851852853854855856857858859860861
Acknowledgments

We would like to thank Stephanie DeYoung and Scott Michaels ofthe Indiana University Yeast 2-Hybrid Facility for assistance with theyeast two-hybrid screens, the Bloomington Drosophila Stock Centerfor fly strains, the Developmental Studies Hybridoma Bank forantibodies and Theresa Shippy for the Tribolium Tsh/Tio cDNA.


ED P

RO

OF

References

Aldaz, S., Morata, G., Azpiazu, N., 2003. The Pax-homeobox gene eyegone is involved inthe subdivision of the thorax of Drosophila. Development 130, 4473–4482.

Alexandre, E., Graba, Y., Fasano, L., Gallet, A., Perrin, L., De Zulueta, P., Pradel, J., Ker-ridge, S., Jacq, B., 1996. The Drosophila teashirt homeotic protein is a DNA-binding protein and modulo, a HOM-C regulated modifier of variegation, is a likelycandidate for being a direct target gene. Mech. Dev. 59, 191–204.

Andrew, D.J., Horner, M.A., Petitt, M.G., Smolik, S.M., Scott, M.P., 1994. Setting limits onhomeotic gene function: restraint of Sex combs reduced activity by teashirt andother homeotic genes. EMBO J. 13, 1132–1144.

Bessa, J., Carmona, L., Casares, F., 2009. Zinc-finger paralogues tsh and tio are function-ally equivalent during imaginal development in Drosophila and maintain their ex-pression levels through auto- and cross-negative feedback loops. Dev. Dyn. 238,19–28.

Bessa, J., Gebelein, B., Pichaud, F., Casares, F., Mann, R.S., 2002. Combinatorial control ofDrosophila eye development by eyeless, homothorax, and teashirt. Genes Dev. 16,2415–2427.

Bischof, J., Basler, K., 2008. Recombinases and their use in gene activation, gene inacti-vation, and transgenesis. Meth. Mol. Biol. 420, 175–195.

Bischof, J., Maeda, R.K., Hediger, M., Karch, F., Basler, K., 2007. An optimized transgen-esis system for Drosophila using germ-line-specific phiC31 integrases. Proc. Natl.Acad. Sci. U. S. A. 104, 3312–3317.

Bonini, N.M., Bui, Q.T., Gray-Board, G.L., Warrick, J.M., 1997. The Drosophila eyes absentgene directs ectopic eye formation in a pathway conserved between flies and ver-tebrates. Development 124, 4819–4826.

Bonini, N.M., Leiserson, W.M., Benzer, S., 1993. The eyes absent gene: genetic control ofcell survival and differentiation in the developing Drosophila eye. Cell 72, 379–395.

Braid, L.R., Verheyen, E.M., 2008. Drosophila nemo promotes eye specification directedby the retinal determination gene network. Genetics 180, 283–299.

Brown, N.L., Sattler, C.A., Paddock, S.W., Carroll, S.B., 1995. Hairy and emc negativelyregulate morphogenetic furrow progression in the Drosophila eye. Cell 80,879–887.

Caubit, X., Core, N., Boned, A., Kerridge, S., Djabali, M., Fasano, L., 2000. Vertebrateorthologues of the Drosophila region-specific patterning gene teashirt. Mech.Dev. 91, 445–448.

Cheyette, B.N., Green, P.J., Martin, K., Garren, H., Hartenstein, V., Zipursky, S.L., 1994.The Drosophila sine oculis locus encodes a homeodomain-containing protein re-quired for the development of the entire visual system. Neuron 12, 977–996.

Choi, K.W., Benzer, S., 1994. Rotation of photoreceptor clusters in the developing Dro-sophila eye requires the nemo gene. Cell 78, 125–136.

Coiffier, D., Charroux, B., Kerridge, S., 2008. Common functions of central and posterior Hoxgenes for the repression of head in the trunk ofDrosophila. Development 135, 291–300.

Core, N., Charroux, B., McCormick, A., Vola, C., Fasano, L., Scott, M.P., Kerridge, S., 1997.Transcriptional regulation of the Drosophila homeotic gene teashirt by the home-odomain protein Fushi tarazu. Mech. Dev. 68, 157–172.

Curtiss, J., Burnett, M., Mlodzik, M., 2007. distal antenna and distal antenna-relatedfunction in the retinal determination network during eye development in Dro-sophila. Dev. Biol. 306, 685–702.

Czerny, T., Halder, G., Kloter, U., Souabni, A., Gehring, W.J., Busslinger, M., 1999. twin ofeyeless, a second Pax-6 gene of Drosophila, acts upstream of eyeless in the controlof eye development. Mol. Cell. 3, 297–307.

Datta, R.R., Cruickshank, T., Kumar, J.P., 2011. Differential selection within the Drosoph-ila retinal determination network and evidence for functional divergence betweenparalog pairs. Evol. Dev. 13, 58–71.

Datta, R.R., Lurye, J.M., Kumar, J.P., 2009. Restriction of ectopic eye formation by Dro-sophila teashirt and tiptop to the developing antenna. Dev. Dyn. 238, 2202–2210.

Erkner, A., Gallet, A., Angelats, C., Fasano, L., Kerridge, S., 1999. The role of Teashirt inproximal leg development in Drosophila: ectopic Teashirt expression reveals dif-ferent cell behaviours in ventral and dorsal domains. Dev. Biol. 215, 221–232.

Erkner, A., Roure, A., Charroux, B., Delaage, M., Holway, N., Core, N., Vola, C., Angelats, C.,Pages, F., Fasano, L., et al., 2002. Grunge, related to human Atrophin-like proteins,has multiple functions in Drosophila development. Development 129, 1119–1129.

Fasano, L., Roder, L., Core, N., Alexandre, E., Vola, C., Jacq, B., Kerridge, S., 1991. The geneteashirt is required for the development of Drosophila embryonic trunk segmentsand encodes a protein with widely spaced zinc finger motifs. Cell 64, 63–79.

Gonzales-Crespo, S., Abu-Shaar, M., Torres, M., Martinez-A, C., Mann, R.S., Morata, G.,1998. Antagonism between extradenticle function and hedgehog signalling inthe developing limb. Nature 394, 196–200.

Hasson, P., Muller, B., Basler, K., Paroush, Z., 2001. Brinker requires two corepressors formaximal and versatile repression in Dpp signalling. EMBO J. 20, 5725–5736.

Henderson, K.D., Isaac, D.D., Andrew, D.J., 1999. Cell fate specification in the Drosophilasalivary gland: the integration of homeotic gene function with the DPP signalingcascade. Dev. Biol. 205, 10–21.

Jang, C.C., Chao, J.L., Jones, N., Yao, L.C., Bessarab, D.A., Kuo, Y.M., Jun, S., Desplan, C.,Beckendorf, S.K., Sun, Y.H., 2003. Two Pax genes, eye gone and eyeless, act cooper-atively in promoting Drosophila eye development. Development 130, 2939–2951.

Keller, S.A., Mao, Y., Struffi, P., Margulies, C., Yurk, C.E., Anderson, A.R., Amey, R.L.,Moore, S., Ebels, J.M., Foley, K., et al., 2000. dCtBP-dependent and -independent re-pression activities of the Drosophila Knirps protein. Mol. Cell. Biol. 20, 7247–7258.

Laugier, E., Yang, Z., Fasano, L., Kerridge, S., Vola, C., 2005. A critical role of teashirt forpatterning the ventral epidermis is masked by ectopic expression of tiptop, a para-log of teashirt in Drosophila. Dev. Biol. 283, 446–458.

Lopes, C.S., Casares, F., 2009. hth maintains the pool of eye progenitors and its downre-gulation by Dpp and Hh couples retinal fate acquisition with cell cycle exit. Dev.Biol. 339, 78–88.



862863864865866867868869870871872873874875876877878879880881882883884885886887888889890891892893894895896897898899

900901902903904905906907908909910911912913914915916917918919920921922923924925926927928929930931932933934935936937

939


Manfroid, I., Caubit, X., Kerridge, S., Fasano, L., 2004. Three putative murine Teashirtorthologues specify trunk structures in Drosophila in the same way as the Dro-sophila teashirt gene. Development 131, 1065–1073.

Mardon,G., Solomon,N.M., Rubin, G.M., 1994. dachshundencodes anuclear protein requiredfor normal eye and leg development in Drosophila. Development 120, 3473–3486.

Mathies, L.D., Kerridge, S., Scott, M.P., 1994. Role of the teashirt gene in Drosophila mid-gut morphogenesis: secreted proteins mediate the action of homeotic genes. De-velopment 120, 2799–2809.

McCormick, A., Core, N., Kerridge, S., Scott, M.P., 1995. Homeotic response elements aretightly linked to tissue-specific elements in a transcriptional enhancer of the tea-shirt gene. Development 121, 2799–2812.

Molloy, D.P., Barral, P.M., Bremner, K.H., Gallimore, P.H., Grand, R.J., 2001. Structural de-terminants outside the PXDLS sequence affect the interaction of adenovirus E1A, C-terminal interacting protein and Drosophila repressors with C-terminal bindingprotein. Biochim. Biophys. Acta 1546, 55–70.

Nibu, Y., Levine, M.S., 2001. CtBP-dependent activities of the short-range Giant repres-sor in the Drosophila embryo. Proc. Natl. Acad. Sci. U. S. A. 98, 6204–6208.

Nibu, Y., Zhang, H., Bajor, E., Barolo, S., Small, S., Levine, M., 1998a. dCtBP mediates tran-scriptional repression by Knirps, Kruppel and Snail in the Drosophila embryo.EMBO J. 17, 7009–7020.

Nibu, Y., Zhang, H., Levine, M., 1998b. Interaction of short-range repressors with Dro-sophila CtBP in the embryo. Science 280, 101–104.

Pai, C.Y., Kuo, T.S., Jaw, T.J., Kurant, E., Chen, C.T., Bessarab, D.A., Salzberg, A., Sun, Y.H.,1997. The Homothorax homeoprotein activates the nuclear localization of anotherhomeoprotein, extradenticle, and suppresses eye development in Drosophila.Genes Dev. 12, 435–446.

Pan, D., Rubin, G.M., 1998. Targeted expression of teashirt induces ectopic eyes in Dro-sophila. Proc. Natl. Acad. Sci. U. S. A. 95, 15508–15512.

Peng, H.W., Slattery, M., Mann, R.S., 2009. Transcription factor choice in the Hippo sig-naling pathway: homothorax and yorkie regulation of the microRNA bantam in theprogenitor domain of the Drosophila eye imaginal disc. Genes Dev. 23, 2307–2319.

Pichaud, F., Casares, F., 2000. homothorax and iroquois-C genes are required for the es-tablishment of territories within the developing eye disc. Mech. Dev. 96, 15–25.

Quiring, R., Walldorf, U., Kloter, U., Gehring, W.J., 1994. Homology of the eyeless gene ofDrosophila to the Small eye gene in mice and Aniridia in humans [see comments].Science 265, 785–789.

Rauskolb, C., Smith, K.M., Peifer, M.,Wieschaus, E., 1995. extradenticle determines segmen-tal identities throughout Drosophila development. Development 121, 3663–3673.

UNCO

RRECT

938


D P

RO

OF

Roder, L., Vola, C., Kerridge, S., 1992. The role of the teashirt gene in trunk segmentalidentity in Drosophila. Development 115, 1017–1033.

Saller, E., Kelley, A., Bienz, M., 2002. The transcriptional repressor Brinker antagonizesWingless signaling. Genes Dev. 16, 1828–1838.

Seimiya, M., Gehring, W.J., 2000. The Drosophila homeobox gene optix is capable of in-ducing ectopic eyes by an eyeless-independent mechanism. Development 127,1879–1886.

Serikaku, M.A., O'Tousa, J.E., 1994. sine oculis is a homeobox gene required for Dro-sophila visual system development. Genetics 138, 1137–1150.

Shen, W., Mardon, G., 1997. Ectopic eye development in Drosophila induced by direct-ed dachshund expression. Development 124, 45–52.

Shippy, T.D., Tomoyasu, Y., Nie, W., Brown, S.J., Denell, R.E., 2008. Do teashirt familygenes specify trunk identity? Insights from the single tiptop/teashirt homolog ofTribolium castaneum. Dev. Genes. Evol. 218, 141–152.

Singh, A., Kango-Singh, M., Choi, K.W., Sun, Y.H., 2004. Dorso-ventral asymmetric func-tions of teashirt in Drosophila eye development depend on spatial cues providedby early DV patterning genes. Mech. Dev. 121, 365–370.

Singh, A., Kango-Singh, M., Sun, Y.H., 2002. Eye suppression, a novel function of tea-shirt, requires Wingless signaling. Development 129, 4271–4280.

Soanes, K.H., Bell, J.B., 2001. The drosophila aeroplane mutant is caused by an I-element insertion into a tissue-specific teashirt enhancer motif. Genome 44,919–928.

Soanes, K.H., MacKay, J.O., Core, N., Heslip, T., Kerridge, S., Bell, J.B., 2001. Identificationof a regulatory allele of teashirt (tsh) in Drosophila melanogaster that affects winghinge development. An adult-specific tsh enhancer in Drosophila. Mech. Dev. 105,145–151.

Sun, Y.H., Tsai, C.J., Green, M.M., Chao, J.L., Yu, C.T., Jaw, T.J., Yeh, J.Y., Bolshakov, V.N.,1995. White as a reporter gene to detect transcriptional silencers specifyingposition-specific gene expression during Drosophila melanogaster eye develop-ment. Genetics 141, 1075–1086.

Venken, K.J., He, Y., Hoskins, R.A., Bellen, H.J., 2006. P[acman]: a BAC transgenic plat-form for targeted insertion of large DNA fragments in D. melanogaster. Science314, 1747–1751.

Wu, J., Cohen, S.M., 2000. Proximal distal axis formation in the Drosophila leg: distinctfunctions of teashirt and homothorax in the proximal leg. Mech. Dev. 94, 47–56.

Zelhof, A.C., Ghbeish, N., Tsai, C., Evans, R.M., McKeown, M., 1997. A role for ultraspira-cle, the Drosophila RXR, in morphogenetic furrow movement and photoreceptorcluster formation. Development 124, 2499–2506.

E



a dissection of the teashirt and tiptop genes reveals a novel ......oof 1 a dissection of the...

Documents